One of the most significant epidemiologic findings regarding IDDM is that there is more than a 30-fold difference in the incidence of disease across populations, where the rates are highest in Finland and lowest in the oriental populations, such as Korea and Japan. Although the factors associated with the global patterns of IDDM incidence are unknown, it is likely that variations in the prevalence of IDDM susceptibility gene(s) are a major determinant. We have recently confirmed that the presence of an amino acid, other than aspartate, in position 57 (non-Asp-57) of the HLA-DQ beta chain is highly associated with IDDM susceptibility, such that the estimated relative risk of developing the disease was 107 for non-Asp-57 homozygotes. Although this genetic marker is strongly related to the development of IDDM among caucasians, its prevalence or relationship to IDDM in other populations or racial groups is not known. Therefore, the proposed investigation will employ a population-based epidemiologic model to determine the contribution of a specific risk factor (i.e., non-Asp-57 homozygosity) to the incidence of IDDM. The populations selected for the study represent areas of high, medium and low IDDM incidence, with caucasians, blacks and orientals. The following specific aims, formulated as null hypotheses, will therefore be tested: 1.) the prevalence of the DNA polymorphisms of the DQ beta chain gene, which are associated with IDDM susceptibility, do not vary across populations or ethnic groups, 2.) the strength of the associations between these molecular markers and IDDM does not differ across populations, 3.) the geographic and racial variations in the incidence of IDDM across the world are unrelated to differences in the prevalence of the genetic markers of IDDM susceptibility.